MIRA: A Musical Intent Refinement Agent for Aligning Text-to-Music Generation with User Intent
cs.SD, cs.AI
Submitted: 2026-10-07
Updated: 2026-10-07
Code: https://github.com/OpenMOSS/MOSS-Music
Project page: https://mirareview.github.io
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Sources
- MusicLM: Generating Music From Text
- Music Flamingo: Scaling Music Understanding in Audio Language Models
- Audio Flamingo 3: Advancing Audio Intelligence with Fully Open Large Audio Language Models
- ACE-Step 1.5: Pushing the Boundaries of Open-Source Music Generation
- GEMS: Agent-Native Multimodal Generation with Memory and Skills
- GenEval 2: Addressing Benchmark Drift in Text-to-Image Evaluation
- AQAScore: Evaluating Semantic Alignment in Text-to-Audio Generation via Audio Question Answering
- LeVo 2: Stable and Melodious Song Generation via Hierarchical Representation Modeling and Progressive Post-Training
- AnyAudio-Judge: A Dynamic Rubric-Based Benchmark and Evaluator for Audio Instruction Following
- MusicEval: A Generative Music Dataset with Expert Ratings for Automatic Text-to-Music Evaluation
- CMI-RewardBench: Evaluating Music Reward Models with Compositional Multimodal Instruction
- Text2midi-InferAlign: Improving Symbolic Music Generation with Inference-Time Alignment
- SongBench: A Fine-Grained Multi-Aspect Benchmark for Song Quality Assessment
- Qwen2.5-Omni Technical Report
- Qwen3-Omni Technical Report
- YuE: Scaling Open Foundation Models for Long-Form Music Generation
- MuQ-Eval: An Open-Source Per-Sample Quality Metric for AI Music Generation Evaluation
Related papers
- Few-Shot Open-Set Audio Classification via Transductive Prototype Refinement and Class Logit Enhancement
- Spectral Masking and Interpolation Attack (SMIA): A Black-box Adversarial Attack against Voice Authentication and Anti-Spoofing Systems
- Remix the Timbre: Diffusion-Based Style Transfer Across Polyphonic Stems
- AVMeme Exam: A Multimodal Multilingual Multicultural Benchmark for LLMs' Contextual and Cultural Knowledge and Thinking
- SoundWeaver: Compositional Warm-Starting for Text-to-Audio Diffusion Serving
- WASIL: In-the-Wild Arabic Spoken Interactions with LLMs